{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:TJROXMEN7VEZYXJATTW5DKSPFE","short_pith_number":"pith:TJROXMEN","canonical_record":{"source":{"id":"2401.14280","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T16:11:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8dfe051ed140d72da0c03a62397ea00edef3af3721dc5fd0c1b123181e16166b","abstract_canon_sha256":"2625b924faec9598a62df03156176b487d6ea7443fb396cfc482aea88ee67441"},"schema_version":"1.0"},"canonical_sha256":"9a62ebb08dfd499c5d209cedd1aa4f290636d3261c73c79ba14413b67555771c","source":{"kind":"arxiv","id":"2401.14280","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.14280","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"arxiv_version","alias_value":"2401.14280v3","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.14280","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"pith_short_12","alias_value":"TJROXMEN7VEZ","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"pith_short_16","alias_value":"TJROXMEN7VEZYXJA","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"pith_short_8","alias_value":"TJROXMEN","created_at":"2026-07-05T08:35:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:TJROXMEN7VEZYXJATTW5DKSPFE","target":"record","payload":{"canonical_record":{"source":{"id":"2401.14280","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T16:11:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8dfe051ed140d72da0c03a62397ea00edef3af3721dc5fd0c1b123181e16166b","abstract_canon_sha256":"2625b924faec9598a62df03156176b487d6ea7443fb396cfc482aea88ee67441"},"schema_version":"1.0"},"canonical_sha256":"9a62ebb08dfd499c5d209cedd1aa4f290636d3261c73c79ba14413b67555771c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:35:41.013538Z","signature_b64":"ziahdd1bVN/7meYR2Kh2LWfXrWjl2gCu84Uo17DOvR9uERY5GIW29WIs67angXfQY+2Md04C6o8J1hyjDsuQBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a62ebb08dfd499c5d209cedd1aa4f290636d3261c73c79ba14413b67555771c","last_reissued_at":"2026-07-05T08:35:41.013182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:35:41.013182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.14280","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:35:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8fuHL1TfDZ5qHOLqQIxX6bqo/4lgiz1Ej612r5RfxK6N7PuPfslapyM5znsIKZcUrSf5MQdY19RFjHBaBiAbCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T20:25:44.352355Z"},"content_sha256":"48cc2d36f41a268f7c6ecfaec83e6de566798b7c6610a2501a9887921a8769b9","schema_version":"1.0","event_id":"sha256:48cc2d36f41a268f7c6ecfaec83e6de566798b7c6610a2501a9887921a8769b9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:TJROXMEN7VEZYXJATTW5DKSPFE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RomanSetu: Efficiently unlocking multilingual capabilities of Large Language Models via Romanization","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anoop Kunchukuttan, Aswanth Kumar, Jaavid Aktar Husain, Jay Gala, Raj Dabre, Ratish Puduppully, Thanmay Jayakumar","submitted_at":"2024-01-25T16:11:41Z","abstract_excerpt":"This study addresses the challenge of extending Large Language Models (LLMs) to non-English languages that use non-Roman scripts. We propose an approach that utilizes the romanized form of text as an interface for LLMs, hypothesizing that its frequent informal use and shared tokens with English enhance cross-lingual alignment. Our approach involves the continual pretraining of an English LLM like Llama 2 on romanized text of non-English, non-Roman script languages, followed by instruction tuning on romanized data. The results indicate that romanized text not only reduces token fertility by 2x-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.14280","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2401.14280/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:35:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"P+y0+EBjry3ZXm4EueQaf2iRtDgq4smge2fQh8rduE/Iu20uq7CeEOc9+fcLqY8rYqRXWTVOOaJRi6Ix6VTQCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T20:25:44.352733Z"},"content_sha256":"466f9706a489a5818d8f6f4e99c2513f6ba9a74a64f966d57c20cfaebc4ca48c","schema_version":"1.0","event_id":"sha256:466f9706a489a5818d8f6f4e99c2513f6ba9a74a64f966d57c20cfaebc4ca48c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TJROXMEN7VEZYXJATTW5DKSPFE/bundle.json","state_url":"https://pith.science/pith/TJROXMEN7VEZYXJATTW5DKSPFE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TJROXMEN7VEZYXJATTW5DKSPFE/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-06T20:25:44Z","links":{"resolver":"https://pith.science/pith/TJROXMEN7VEZYXJATTW5DKSPFE","bundle":"https://pith.science/pith/TJROXMEN7VEZYXJATTW5DKSPFE/bundle.json","state":"https://pith.science/pith/TJROXMEN7VEZYXJATTW5DKSPFE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TJROXMEN7VEZYXJATTW5DKSPFE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:TJROXMEN7VEZYXJATTW5DKSPFE","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2625b924faec9598a62df03156176b487d6ea7443fb396cfc482aea88ee67441","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T16:11:41Z","title_canon_sha256":"8dfe051ed140d72da0c03a62397ea00edef3af3721dc5fd0c1b123181e16166b"},"schema_version":"1.0","source":{"id":"2401.14280","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.14280","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"arxiv_version","alias_value":"2401.14280v3","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.14280","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"pith_short_12","alias_value":"TJROXMEN7VEZ","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"pith_short_16","alias_value":"TJROXMEN7VEZYXJA","created_at":"2026-07-05T08:35:41Z"},{"alias_kind":"pith_short_8","alias_value":"TJROXMEN","created_at":"2026-07-05T08:35:41Z"}],"graph_snapshots":[{"event_id":"sha256:466f9706a489a5818d8f6f4e99c2513f6ba9a74a64f966d57c20cfaebc4ca48c","target":"graph","created_at":"2026-07-05T08:35:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2401.14280/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This study addresses the challenge of extending Large Language Models (LLMs) to non-English languages that use non-Roman scripts. We propose an approach that utilizes the romanized form of text as an interface for LLMs, hypothesizing that its frequent informal use and shared tokens with English enhance cross-lingual alignment. Our approach involves the continual pretraining of an English LLM like Llama 2 on romanized text of non-English, non-Roman script languages, followed by instruction tuning on romanized data. The results indicate that romanized text not only reduces token fertility by 2x-","authors_text":"Anoop Kunchukuttan, Aswanth Kumar, Jaavid Aktar Husain, Jay Gala, Raj Dabre, Ratish Puduppully, Thanmay Jayakumar","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T16:11:41Z","title":"RomanSetu: Efficiently unlocking multilingual capabilities of Large Language Models via Romanization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.14280","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:48cc2d36f41a268f7c6ecfaec83e6de566798b7c6610a2501a9887921a8769b9","target":"record","created_at":"2026-07-05T08:35:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2625b924faec9598a62df03156176b487d6ea7443fb396cfc482aea88ee67441","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T16:11:41Z","title_canon_sha256":"8dfe051ed140d72da0c03a62397ea00edef3af3721dc5fd0c1b123181e16166b"},"schema_version":"1.0","source":{"id":"2401.14280","kind":"arxiv","version":3}},"canonical_sha256":"9a62ebb08dfd499c5d209cedd1aa4f290636d3261c73c79ba14413b67555771c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9a62ebb08dfd499c5d209cedd1aa4f290636d3261c73c79ba14413b67555771c","first_computed_at":"2026-07-05T08:35:41.013182Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:35:41.013182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ziahdd1bVN/7meYR2Kh2LWfXrWjl2gCu84Uo17DOvR9uERY5GIW29WIs67angXfQY+2Md04C6o8J1hyjDsuQBw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:35:41.013538Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.14280","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:48cc2d36f41a268f7c6ecfaec83e6de566798b7c6610a2501a9887921a8769b9","sha256:466f9706a489a5818d8f6f4e99c2513f6ba9a74a64f966d57c20cfaebc4ca48c"],"state_sha256":"ed7a9b4447bff93694557e0b8e063157fff7c9f457e9b1e70c9a529290d08b64"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"woJ2ZitQUWZeIEc/qxeN00S3OIblEZn2up1oxzHjI39dQgvoWsv4a+XxJL78gvisApSyKJz+L0vuxtlZAlfXBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T20:25:44.355054Z","bundle_sha256":"1b10e346fb60f1dbb9d950bd858113ea7fdaf70aab05aed906053178626cc5a6"}}